Back to EveryPatent.com



United States Patent 5,525,482
States ,   et al. * June 11, 1996

Method and cell line for testing cytotoxicity and mutagenicity of a chemical

Abstract

A method of testing the cytotoxicity and mutagenicity of a chemical includes the steps of exposing test cells to the chemical in vitro, intracellularly metabolizing the chemical into a mutagenic or cytotoxic metabolite and then detecting gene/protein/cell damage in the test cells as an indication of the mutagenicity/cytotoxicity of the chemical. A cell line is provided for testing cytotoxicity and mutagenicity of the chemicals, the cell line consisting essentially of fibroblasts normally having no detectable cytochrome P450 mixed function oxidase enzyme activity. The fibroblasts are transformed with chimeric gene constructs containing cytochrome P450 coding sequences and have intracellular cytochrome P450 oxidative metabolizing activity.


Inventors: States; J. Christopher (Royal Oak, MI); Hines; Ronald N. (West Bloomfield, MI); Novak; Raymond F. (West Bloomfield, MI)
Assignee: Wayne State University (Detroit, MI)
[*] Notice: The portion of the term of this patent subsequent to January 19, 2010 has been disclaimed.
Appl. No.: 339658
Filed: November 15, 1994

Current U.S. Class: 435/32; 435/29; 435/455
Intern'l Class: C12Q 001/18; C12N 005/10; C12N 015/09
Field of Search: 435/7.21,7.23,29,172.3,240.2,32 536/23.5


References Cited
U.S. Patent Documents
4696898Sep., 1987Fitts et al.435/69.
5164313Nov., 1992Gelboin et al.435/189.
5180666Jan., 1993States et al.435/29.


Other References

McManus et al., Cancer Res., v. 50, Jun./1990, pp. 3367-3376.
Crespi et al., Toxicology, v. 82, 1993, pp. 89-104.
Perkus et al., J. Virol., vol. 63, 1989, pp. 3829-3836.
Hay et al., Mutat. Res., vol. 130, 1984, pp. 321-322.
Webster's Ninth New Collegiate Dictionary, 1990, Merriam-Webster Inc., Springfield, MA, p. 412.
B. Ames et al: Science 176:47-49 (1972).
B. Ames et al: Mutat. Res. 31:347-364 (1975).
A. Akeson et al., J. Biol. Chem. 263:16291-16296 (1988).
V. Bohr et al., Cell 40:359-369 (1985).
J. Cleaver et al., Ann Rev. Genet 9:19-38 (1975).
J. Cleaver et al., Cancer Res. 33:362-369 (1973).
C. Crespi et al., Carcinogenesis 10:295-301 (1989).
R. Davies et al., Carcinogenesis 10:885-891 (1989).
J. Doehmer et al., Proc. Natl. Acad. Sci (USA) 85:5769-5773 (1988).
S. Dogra et al., Molec. Pharmacol. 37:608-613 (1990).
F. Gonzalez et al., Mutation Res. 247:113-127 (1991).
F. Guengerich et al., Biochemistry 21:6019-6030 (1982).
H. Govan III et al., Nucleic Acids Res 18-3823-3830 (1990).
D. Hamer et al., J. Mol. & Appl. Gen. 1:273-288 (1982).
P. Iversen et al., Arch. Biochem. Biophy. 256:397-401 (1987).
K. Kraemer et al., Proc. Natl. Acad. Sci. (USA) 72:59-63 (1975).
W. Kleijer et al., Mutat. Res. 9:517-523 (1970).
A. Klotz et al., Anal. Biochem. 140:138-145 (1984).
R. Lubet et al., Mutat. Res. 142:127-131 (1985).
T. McLemore et al., J. Nat'l. Cancer Inst. 82:1333-1339 (1990).
A. Miller, Anal. Biochem. 133:46-57 (1983).
D. Nebert et al., DNA 6:1-11 (1987).
D. Nebert et al., DNA 8:1-13 (1989).
M. Pasanen et al., Pharmacol & Toxicol. 62:311-317 (1988).
A. Puga et al., DNA and Cell Biol. 9:425-436 (1990).
D. Thomas et al., Proc. Nat'l Acad. Sci. (USA) 85:3723-3727 (1988).
L. Thompson, Mutation Res. 247:213-219 (1991).
J. States, Mechanisms & Cons. of DNA Damage Proc., UCLA Symposia on Molecular & Cellular Biology, New Series, 83:307-311 (1988).
J. van Gurp et al., Mol. Cell Biol. 4:1597-1604 (1984).

Primary Examiner: Schwartz; Richard A.
Assistant Examiner: Ketter; James
Attorney, Agent or Firm: Reising, Ethington, Barnard and Perry

Parent Case Text



This is a continuation of application Ser. No. 07/990,295 filed on Dec. 9, 1992, abandoned, which is a CIP of 07/721,775, filed Jun. 27, 1991, U.S. Pat. No. 5,180,666.
Claims



What is claimed is:

1. A method of testing for cytotoxicity to humans of a chemical comprising the steps of:

(a) exposing human fibroblast test cells normally not including any cytochrome P450 activity to potentially toxic chemicals, the test cells having been transformed to express cytochrome P450 enzymes, under the control of a controllable promoter through the human CYP1A1 gene, upon exposure to the chemical in vitro;

(b) intracellularly metabolizing the chemical into a cytotoxic metabolite by oxidatively metabolizing the chemical within the fibroblasts through the intracellular cytochrome P450 mixed function oxidase enzymes expressed by said human fibroblast test cells; and (c) detecting gene damage in the test cells as an indication of cytotoxicity of the chemical.

2. A method as set forth in claim 1 wherein the human fibroblast test cells of step (a) comprise a first test group of cells having the metabolic activity of said human fibroblast test cells, and control cells genetically equivalent to the first test group of cells but having significantly less metabolic activity.

3. A method as set forth in claim 2 wherein with said human fibroblast test cells of said first test group are created by inducing expression of enzymes having the metabolic activity of said first test group.

4. A method as set forth in claim 3 wherein the said human fibroblast test cells of said first test group are created by inducing expression of intracellular cytochrome P450 mixed function oxidase enzymes.

5. A method as set forth in claim 4 wherein said human fibroblast test cells have been transformed to express cytochrome P450 enzymes under control of the mouse metallothionein-1 promoter and placed in a metal containing media, said human fibroblast test cells of said first test group are created by modulating the metal ion content of the media within ranges which are nontoxic to the cells to modulate the expression of the P450 enzyme activity.

6. A method as set forth in claim 1 wherein the human fibroblast test cells are DNA repair deficient thereby providing cells sensitive to cytotoxicity at low doses as compared to doses, required for detectable cytotoxicity in cells which are not DNA repair deficient.

7. A method of making a cell line for testing mutagenicity of a chemical comprising the step of transforming a DNA repair deficient human XPA fibroblast cell line to express cytochrome P450 enzyme activity through the human CYP1A1 gene intracellularly.
Description



TECHNICAL FIELD

The present invention relates to in vitro test systems used for assessing cytotoxicity and genotoxicity of chemicals. More specifically, the present invention relates to in vitro test systems which replace whole animal models currently used for such tests and is particularly useful for testing the cytotoxicity and genotoxicity of chemicals, the toxicity of the chemicals requiring metabolic activation.

BACKGROUND ART

It is desirable to develop non-animal models for assessing cytotoxicity and genotoxicity of chemicals in humans. Accordingly, a human cell culture system offers the advantage of species identity and eliminates the need to extrapolate results from experiments in rodent or bacterial cells.

Metabolic activation is essential for the conversion of a wide variety of chemicals to the ultimate cytotoxic or genotoxic agents. It is therefore desirable to provide a toxicological test system which is proficient in bioactivation of xenobiotics in order to reveal the toxicologic potential of an agent. Cytochromes P450 comprise the most important class of enzymes that bioactivate xenobiotics.

Cytochromes P450 are implicated in the activation of many cytotoxic agents, pro-mutagens and cancer chemotherapy agents. In particular, cytochrome P450IA1 (CYP1A1) has been implicated in lung cancer by virtue of its ability to bioactivate components of cigarette smoke such as benzo(a)pyrene into potent mutagenic agents.

The prokaryotic test system most commonly used to detect mutagenicity at the present time is the Ames test (1). The Ames test is limited by the fact that bacteria used in the test lack bioactivating enzymes such as the cytochromes P450. Therefore the system cannot detect the mutagenicity of agents requiring metabolic activation.

A modified Ames test (2) has been developed which includes premetabolism by microsomal preparations. This test suffers from the fact that many agents, once metabolized to their active forms, are unable to cross the plasma membrane to enter the cells. In other words, the extracellular location of the enzymes greatly prejudices the effectiveness of the test. It is therefore further desirable to provide a test system consisting of mammalian cells metabolically competent for bioactivation.

There have been many approaches used by researchers to develop cell lines for use as test systems for bioactivation of chemicals. Each of these approaches are deficient in at least one aspect if not many aspects. With regard to these type of tests, there are three important considerations for developing engineered mammalian cell lines for examining bioactivation of chemicals and assessing their toxologic potential.

The first consideration is appropriate and stable expression of the chimeric gene construct. For example, Chinese hamster ovary cells have been transformed with rat cytochrome P450 IA1 (CYP1A1,(3)) and IIB1 (CYP2B1,(4)) cDNAs in an expression vector. The expression of the P450 cDNA in the vector is controlled by the strong, constitutive SV40 early promotor. One cell line had a few copies of the CYP1A1 but the other two had many copies. One of the high copy transformants produced several bands on genomic Southern blot strongly hybridizing with CYP1A1 probe suggesting rearrangement of the integrated genes. Multi-copy integration events are often unstable and the multi-copy CYP1A1 transgenes in these CHO lines are no exception (3). One of the lines is losing copies and the very high copy number transformant appears to be amplifying the transgene. Only one of the three is reported to be relatively stable. Only the relatively stable transformant and the low copy transformant show increased mutagenicity upon exposure to benzo(a)pyrene (BAP) and BAP-(trans)-7,8-diol, two compounds which are mutagenic only after metabolism by CYP1A1.

These studies show that expression of CYP1A1 or CYP2B1 resulted in an increased mutagenicity after exposure of the cells to compounds known to be bioactivated by these enzymes. However, the number of transformants obtained which remain stable in their expression of the P450 transgenes is few. It is hypothesized that this is due to the promotor used in the expression vector being a strong constitutive promotor in most situations. Constant, high levels of P450 expression can be deleterious to the cells. It is more desirable to have a P450 cDNA under control of an inducible promotor of moderate strength as opposed to the strong constitutive promotor used in the prior art.

The second consideration is the ability to draw conclusions about dose level in the human situation. One aspect of the problem is that human cells are more complete in their capability for DNA repair than are rodent cells which concentrate their repair efforts on transcriptionally active regions of the chromatin. Transformation of human cells with P450 cDNAs in expression vectors is more desirable for drawing conclusions about mutagenic doses in humans.

Other researchers have initiated studies of transfected cytochrome P450 cDNAs in human lymphoblasts (4,5,6). This system addresses the species specificity problem but does not allow for inducible expression of P450 as does the vector used to transform human cells in accordance with the present invention.

Very recent research has been directed to applying a mutant hamster cell line deficient in DNA repair to problems in environmental mutagenesis by introducing the mouse cytochrome P.sub.3 450 (P4501A2 subfamily) gene for metabolic activation of aromatic amines (6a). Others suggest an in vitro mutagenic system based on activation by human P450s to supplement other test systems (6b).

A third consideration is the ability to assess gene specific DNA damage. Comparison of DNA damage in specific genes is facilitated by using cells which do not repair any of the damage sites. For this reason, it is desirable to use human DNA excision repair deficient cells in addition to normal DNA repair proficient cells.

SUMMARY OF THE INVENTION

In accordance with the present invention, there is provided a method of testing for cytotoxicity and mutagenicity of a chemical by exposing test cells to the chemical in vitro, intracellularly metabolizing the chemical into a mutagenic and/or cytotoxic metabolite, and detecting cell death and gene damage in the test cells as an indication of cytotoxicity and mutagenicity of the chemical.

The present invention further provides a cell line, for testing cytotoxicity and mutagenicity by a chemical consisting essentially of fibroblasts which normally have no detectable cytochrome P450 mixed function oxidase enzyme activity, the fibroblasts being transformed with chimeric gene constructs containing cytochrome P450 coding sequences and having intracellular cytochrome P450 oxidative metabolizing activity.

FIGURES IN THE DRAWINGS

Other advantages of the present invention will be readily appreciated as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein:

FIG. 1 is a map of the pRNH127 construct;

FIG. 2 is a map of the pRNH155 construct;

FIG. 3A-C is the DNA sequence of construct pRNH127 (SEQ. ID No. 1);

FIG. 3D-F is the DNA sequence of pRNH155 (SEQ. ID No. 1);

FIG. 4 is a partial map of the 5830 bp human CYP1A1 gene;

FIG. 5 shows induction of CYP1A1 detected by Western blot in HEPG2 cells but not in XPA fibroblasts;

FIG. 6 shows induction of CYP1A1 mRNA detected by Northern blot in HEPG2 cells but not in XPA fibroblasts;

FIG. 7 shows test results of synthesized EFEE with HepG2 cells as a chromatograph developed with ethylacetate: benzene and visualized by illumination with long wave ultraviolet light;

FIG. 8 is a chromatograph visualized by illumination with long wave ultraviolet light showing analysis of NEO+transformants for CYP1A1 activity inducible by 5 .mu.M CdSO.sub.4 ;

FIG. 9 shows plated cells demonstrating benzo(a)pyrene-trans-7,8-diol (BPD) is cytotoxic only in CYP1A1 expressing cells;

FIG. 10 shows plated cells demonstrating the prevention of BPD cytotoxicity by (.alpha.-NF); and

FIG. 11 shows plated cells demonstrating that benzo (a) pyrene-trans-7,8-diol-9,10-epoxide is equally cytotoxic in all cell lines tested.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a method of testing for cytotoxicity and mutagenicity of a chemical, the method generally including the steps of exposing test cells to the chemical in vitro, intracellularly metabolizing the chemical into a cytotoxic and/or mutagenic metabolite, and detecting cell death and/or gene damage in the test cells as an indication of cytotoxicity and/or mutagenicity of the chemical. The invention differs from existing technology by the engineering of cells such as human fibroblasts with an inducible chimeric gene for expressing enzymes which metabolize chemicals. Unlike prior art technologies, such as the Ames test, the present invention tests for cytotoxicity and mutagenicity by utilizing human cells containing chemical metabolizing enzymes whereas the most commonly used test used prokaryotic cells which need to be supplemented extracellularly with the enzymes active in the metabolism of the chemicals. Moreover, the eukaryotic prior art systems include cells containing protective enzymes which can detoxify the metabolized chemicals thereby resulting in a false negative mutagenicity test. Hence, the present test method provides a balance between metabolic activation, detoxification and toxicity/mutagenicity which approximates those inherent in normal human cells.

The invention provides a two stage protocol. First, cytotoxicity of a chemical can be determined. If found, then mutagenicity (mechanism of action) can be determined.

The preferred test cell used in accordance with the present invention are cells which normally have no detectable cytochrome P450 mixed function oxidase enzyme activity. Examples of such cells are human skin fibroblasts derived from xeroderma pigmentosum group A (XPA) and DNA repair normal donors, and human epithelial cells. The human fibroblast cells are preferred because fibroblasts are adherent and readily form colonies which are easy to count in a test system. Unlike prior art systems which use human lymphoid cells which are nonadherent to a test plate and therefore cause difficulties in counting surviving colonies in a test system, the present invention preferably utilizes fibroblasts which are adherent to a test plate and form colonies which are easily counted. More specifically, a common system for detection of mutagenicity includes a plating step wherein cells are grown on plates in media in a dose response test to a chemical. Surviving cells are counted in media which is selective for the mutant thereby indicating the relative mutagenicity of a chemical in a dose response test. (Dose response test meaning different plates containing cells are exposed to increasing concentrations of the chemical). Increased cell survival of the mutated cells which are the only cells that survive in the test media are an indication of response to chemical, this being shown in a dose dependent manner.

As stated above, test cells are used which in their normal state do not include intracellular cytochrome P450 mixed function oxidase enzymes. The test cells are transformed to express the chemical metabolizing enzymes, preferably including an inducible promotor. That is, the test cells are preferably fibroblasts normally not including any cytochrome P450 oxidative activity. The test cells are transformed to express cytochrome P450 enzymes such that the metabolizing step discussed above is more specifically defined as oxidatively metabolizing the chemical within the fibroblasts through the intracellular cytochrome P450 mixed function oxidase enzyme expressed by the transformed fibroblasts.

Preferably, the human XPA fibroblasts are transformed by the introduction of a chimeric cytochrome P450IA1 (CYP1A1) expression construct. FIGS. 1 and 2 show maps of the pRNH127 construct, a 6387 bp construct, and the pRNH155 construct, respectively, the latter a 6387 bp construct. FIGS. 3a and b show the base sequence for the pRNH127 and pRNH155 constructs, respectively.

The pRNH127 construct is a mouse metallothionein (MMT-1) CYP1A1 expression vector that was constructed by first isolating the MMT-1 promotor from p13MMT as an EcoRI/HinDIII fragment, and trimming the ends with HaeIII and Sau3AI. This fragment was force cloned into pRNH62 that had been digested with HinDIII, treated with T4 DNA polymerase to prepare blunt-ends and then digested with BclI (thereby eliminating intron 1 sequences).

pRNH155 is the same clone that was constructed using a different strategy. A MMT-1 CYP1A1 expression vector was constructed by first isolating the MMT-1 promotor from p13MMT as a 467 base pair Sau3AI fragment which was then cloned into pRNH1 that had been digested with BclI. After verifying the correct orientation of the promotor with EcoRI, CYP1A1 intron 1 sequences were removed by digesting to completion with HinDIII, followed by a partial EcoRI digest. The 5491 bp band representing the desired deletion was isolated, treated with T4 DNA polymerase to repair the ends, and then ligated to close the plasmid. The 3' end of the CYP1A1 gene was isolated from pRNH59 as an 895 bp EcoRI fragment and then cloned into the now unique EcoRI site of the intermediate plasmid. Orientation was verified by digestion by BamHI/BglII.

FIG. 4 shows a partial map of the 5830 base pair human CYP1A1 gene. In the Figure, the open boxes represent non-coding exon information and the closed boxes represent coding exon information. The position of restriction endonuclease recognition sites are shown for various subclones is shown below the map. The BclI/EcoRI fragment containing CYP1A1 coding information cloned into expression vectors to be used is denoted by the broken line.

The present invention preferably uses fibroblasts transformed to express the cytochrome P4501A1 mixed function oxidase enzyme (ATCC Reference CRLll172 and CRL 11173). The cytochrome P450 mixed function oxidase enzyme family catalyzes the oxidative metabolism of a variety of xenobiotics in mammalian cells including environmental protoxicants, steroid hormones, cancer chemotherapeutic agents, and procarcinogens. As a result of cytochrome P450-mediated metabolism, certain compounds are converted to reactive intermediates which are capable of damaging cellular macromolecules. The mammalian cytochrome P450 gene superfamily includes at least eight gene families (7-9). Experiments on rat and liver have demonstrated that some of the cytochrome P450 gene families produce at least 20 separate cytochrome P450 enzyme forms. The tissue specificity of cytotoxicity and carcinogenesis caused by some xenobiotics may be related to the tissue specific expression of cytochromes P450 which activate those compounds.

Because of this relationship of the cytochrome P450 to chemical activation of procarcinogens to carcinogens, applicant chose the P450 system to be incorporated into the test cells. Applicant specifically chose the CYP1A1 activity as this gene is well characterized. In particular, cytochrome P450IA1 has been implicated in lung cancer by virtue of its ability to bioactivate components of cigarette smoke such as benzo(a)pyrene into potent cytotoxic and mutagenic/carcinogenic agents.

As discussed above, the construct used preferably in accordance with the present invention includes the mouse metallothionein promotor (MMT-1). Unlike prior art systems which utilize the strong constitutive SV40 early promotor, the present invention preferably uses an inducible promotor of moderate strength. Preferably, the MMT-1 promotor is easily induced by exposure of cells to nontoxic levels of metals such as zinc or cadmium. Accordingly, control cells can be made directly in accordance with the present invention but are not exposed to the inducible levels of metals whereas test cells are exposed to metals such as zinc or cadmium to induce expression of the CYP1A1 activity. Since the MMT-1 promotor only requires nontoxic levels of the metals, this being previously well characterized (9a), the test cells include an internal control mechanism which has no effect on the cells which would alter test results.

A further advantage of using fibroblast cells such as the XPA cells is the fact that the XPA cells are produced from a human autosomal recessive genetic disorder caused by the inability to repair DNA damage caused by ultraviolet light and chemicals which form bulky DNA adducts (10-12). XPA cells are capable of repairing other types of DNA damage such as single strand breaks caused by ionizing radiation (13). The genetics of xeroderma pigmentosum (XP) have been defined unusually well by somatic cell fusion studies using fibroblast lines established from approximately 100 individuals with XP disorder. Members of XP complementation group A (XPA) exhibit the lowest DNA excision repair capacity (14) and therefore are most sensitive to DNA damage. Thus, these cells provide the most sensitivity for detection of agents which exert cytotoxic and genotoxic effects by bulky adduct damage to DNA.

Various methods have been used for the detection of gene damage in test cells as an indication of mutagenicity of a chemical. For example, gene specific DNA damage resulting from UV exposure has been monitored by an elaborate method making use of bacterial or bacteriophage pyrimidine dimer specific endonucleases and Southern blotting of DNA treated with these endonucleases and separated in denaturing gels (15). An analogous procedure using the E. coli ABC-excision nuclease has been useful in analyzing large adduct DNA damage (16). Other methods to measure large adduct DNA damage have recently been developed (17). This latest method exploits the inability of Taq DNA polymerase to utilize DNA containing large adduct damage as a template for DNA synthesis. A consequence of this phenomenon is that polymerase chain reaction (PCR) amplification of a region containing such damage is reduced proportionately to the number of damaged segments in the sample. One can use the decrease in PCR amplification to measure the amount of damage sustained by a specific DNA segment in a population of cells. Additionally, by performing the assay at various time points after exposure to a damaging agent, the rate of repair of a specific segment can also be monitored and compared to the rate of repair of other segments. This is a powerful assay which can be performed relatively quickly and easily.

Preferably the present invention utilizes oligonucleotides flanking the mutational hot spots in the p53 gene as primers for PCR assay of gene specific DNA damage. These oligonucleotides have been used by others to amplify these regions of the p53 gene (18). This method produces fragments of appropriate size for the DNA damage assay detailed below. Similarly, oligonucleotides for the ras gene are used to monitor DNA damage in that gene as discussed below.

More specifically, the following assays are used to test for gene specific DNA damage, mutagenicity, and cytoxicity.

a. Gene specific DNA damage assay This assay is adapted from Govan et al, Nucleic Acids Research 18:3823-3830 (1990):

Quantitative polymerase chain reaction (PCR) amplification is performed by limiting the amount of template DNA such that the yield of product is linearly proportional to input DNA levels. This normally is achieved in the range of 10 to 200 ng input genomic DNA. In order to quantitate the product, 0.1 .mu.Ci .alpha.-.sup.32 P-dCTP is added to the reaction such that the product will be radiolabelled. PCR products are fractionated by gel electrophoresis and visualized by staining in 0.1 .mu.M ethidium bromide. The gel is dried and autoradiographed by exposing it to X-ray film. The autoradiogram is used to localize the labelled product which is then excised and the incorporated .sup.32 p is measured by scintillation spectrometry. This serves as a measure of product formation.

b. Mutagenicity assay:

Mutations in the hypoxanthine-phosphoribosyl-transferase (HPRT) gene induced by exposure of cells to chemicals are quantified in a dose-response experiment and used as a measure of mutagenicity. The HPRT gene is on the X chromosome and therefore, only one copy is active in mammalian cells. This provides an appropriate single copy gene for mutation studies.

The assay is performed on cells that are induced for expression of CYP1A1 and on control uninduced cells. After incubation plus (induced) and minus (uninduced) metal ions, the cells are exposed to various doses of chemical (i.e. 1 to 50 .mu.M benzo(a)pyrene) for a set time period. After exposure to chemical, the cells are washed and harvested and then replated at lower density in media containing 6-thioguanine (7 .mu.g/ml) to select for cells with mutated HPRT gene. Colonies which form from the growth of single cells with mutated HPRT genes are counted. The colony numbers serve as the data points. If a chemical, such as benzo(a)pyrene, is metabolized to a mutagenic compound by CYP1A1, many more colonies will result at any given does from the cells induced for expression of CYP1A1.

c. Cytotoxicity assay:

The cytotoxicity assay is similar to the mutagenicity assay except that non-selective media is used. Cells induced or uninduced for CYP1A1 expression and exposed to various doses of chemical are washed, harvested and replated at lower density in normal media and allowed to form colonies. Cytotoxicity is indicated by a decrease in the number of colonies with increasing dose of chemical. Metabolic activation of chemical by CYP1A1 is indicated by a further decrease in number of colonies formed at a given dose by cells induced for CYP1A1 expression.

EXPERIMENTATION

The following experiments provide the basis for using human cell lines developed as a suitable nonanimal substitute for the determination of the cytotoxicity, genotoxicity, and potential carcinogenicity of a variety of compounds.

MATERIALS AND METHODS

a. Preparation of the CYP1A1 expression construct:

CYP1A1 coding and polyadenylation sequences contained in plasmid pRNH62 (a genomic subclone, FIG. 4) was isolated and cloned into pUC8 (a cloning vector with a multiple cloning site) using standard recombinant DNA techniques. The mouse metallothionein-I (MMT-I) promoter fragment (containing one metal responsive element) was inserted into the pUC8-CYP1A1 clone in the proper orientation to direct the transcription of CYP1A1 sequences. This construct is co-transfected with a dominant selectable marker (pRSV-NEO) into both SV40 transformed XPA fibroblasts (GM4429 & GM4312) and DNA repair proficient SV40 transformed human fibroblasts (GM0637).

Single colonies of transfected cells are selected by incubation in media containing G-418. These colonies are then analyzed for constitutive and inducible expression of CYP1A1. The MMT-I promoter more appropriately approximates the expression levels of P450s in non-hepatic tissues. Other human P450 cDNAs obtained by RT-PCR from human liver mRNA (or PCR of human liver cDNA library DNA) can be inserted into the expression vector for transfection of human cells.

b. Transfections and selection of transformants:

The XPA and normal fibroblasts have been obtained from the Genetic Mutant Cell Respository, Camden, N.J. The cell lines are SV40 transformed which is essential for them to be transfected efficiently and to be propagated eternally after transformation. The cell lines used are: XPA, GM4429 and GM4312; normal, GM0637. These cell lines are well characterized as to their DNA repair capabilities having been used by many laboratories over the years. The XPA and normal fibroblasts are transfected with Ca-PO.sub.4 co-precipitates (19) of the CYP1A1 expression construct and pRSV-NEO (the NEO cartridge under control of the Rous sarcoma virus long terminal repeat). Several ratios (1:1, 5:1 & 10:1) of CYP1A1:pRSV-NEO DNA are used to increase the likelihood of obtaining transformants with a single copy, functional CYP1A1 chimeric gene construct. Transformants are selected by incubation in medium containing Geneticin (300 .mu.g/ml).

Single colonies are isolated with cloning rings and propagated individually in media containing Geneticin at 75 .mu.g/ml to assure maintenance of the transgenes. Only one colony is picked from a single plate to ensure that duplicate clones will not be picked. The inducibility of CYP1A1 is assayed in 24 well dishes with ethoxyfluorescein ethyl ester (EFEE) as substrate (21). This assay can be performed on intact cells without cytotoxicity by overlaying the cell monolayer with phosphate buffered saline (PBS) containing EFEE. The cells are incubated for two hours and the PBS containing the EFEE and its metabolites is decanted, acidified and analyzed by thin layer chromatography. EFEE and fluorescein ethyl ester (FEE), the product of CYP1A1 metabolism of EFEE are easily separated and visualized under long wave UV illumination. Thus, CYP1A1 activity in induced and uninduced cells can be compared early after isolation of the clones before substantial propagation has occurred. Uninduced cells produce little or no FEE and induced cells produce ample amounts of FEE. Applicants have synthesized sufficient quantities to perform several thousand assays, verified the quality of the product by testing in HepG2 cells (FIG. 5) and used it to screen a small set of transformants (FIG. 6) with positive results as discussed below.

The copy number of both NEO and CYP1A1 constructs integrated is determined in genomic DNA from each individually propagated colony which shows inducible CYP1A1 activity. It is expected to obtain several colonies which fit the criteria of a single inducible copy of CYP1A1. This will provide several lines in which to test the xenobiotic activities. Thus, there is control for the potential variability which may be due to integration position of the transgenes.

c. Characterization of transformants:

The transformants are characterized with regard to: 1) copy number of CYP1A1 gene constructs, 2) expression levels of CYP1A1 mRNA in the uninduced state and induced by incubation with heavy metals, and 3) the levels of CYP1A1 protein and enzymatic activity in the uninduced state and induced by incubation with heavy metals which regulate the MMT-I promoter in the expression construct.

The copy number of CYP1A1 expression constructs present in the transgenes of individual transformants is assessed using quantitative PCR or slot blot analyses. Quantitative PCR has been used to quantitate bulky adduct DNA damage (17) and can be adapted easily to quantitate the copy number of integrated sequences. Oligonucleotide primers which specifically direct the amplification of the expression construct are used to amplify integrated sequences from genomic DNA of transformants. One oligonucleotide is derived from the MMT-I promoter sequences and the other derived from within the CYP1A1 coding sequences. Standardization is achieved by amplification of a single copy sequence such as the p53 gene to be analyzed for DNA damage (see below) and by reconstruction using non-transformant genomic DNA samples with CYP1A1 expression construct plasmid DNA added to simulate different copy numbers. Alternatively, slot blot or genomic Southern blot analyses can be used to quantitate the copy number of chimeric gene constructs in each transformant.

Expression of CYP1A1 mRNA is assessed by quantitative RT-PCR (PCR amplification of first strand cDNA prepared from mRNA of transformants) and confirmed by northern blot analyses using CYP1A1 cDNA as hybridization probe in specific instances (transformants to be used for further work). Oligonucleotides from two different exons within the CYP1A1 mRNA are used specifically to direct the amplification of mature mRNA sequences without interference from unexpressed native genes. CYP1A1 mRNA levels are assessed in dose response experiments wherein the expression of the CYP1A1 construct is induced by addition to the media of CdSO.sub.4 from 1 to 10 .mu.M. This is the range over which XPA cells transfected with a MMT-I promoted chloramphenicol acetyl transferase (CAT) gene displayed induction. Measurement of actin mRNA levels serve as internal controls for normalization in both RT-PCR and northern blot analyses.

CYP1A1 enzyme activity is quantitated using the ethoxyresorufin O-deethylase assay, essentially as described by Klotz et al. (22). The specificity of this reaction for the CYP1A1 isozyme has been well established in both animal models (23,24) and human tissue samples (24, 25). In brief, the reaction is carried out in 0.1M KP.sub.i, pH7.8 containing 0.1M NaCl, 2 .mu.M ethoxyresorufin, 5 to 150 .mu.g of microsomal protein and 0.5 mM NADPH. The formation of resorufin is monitored at (.epsilon.572=73 mM.sup.-1 cm.sup.-1). Alternatively, the same assay can be carried out using whole cell homogenate, but with the addition of 50 .mu.M dicumarol to inhibit the reduction of resorufin by cytosolic quinone oxidoreductase (26). Levels of CYP1A1 protein can also be monitored by Western blot using the antibody to rat CYP1A1. The time course of induction also will be monitored to determine the minimum exposure times necessary to achieve induction. In addition the decay of CYP1A1 activity after inducing agents are removed will be determined in order to define the maximum time during which xenobiotics may be metabolized.

d. Dose response to xenobiotics:

Cells are seeded at 10.sup.5 cells per 6 cm dish and half the dishes are incubated in media containing 5 .mu.M CdSO.sub.4 to induce the expression of the CYP1A1 construct. CdSO.sub.4 is removed by washing the cells with phosphate buffered saline containing 5 mM EDTA after a suitable induction period. Xenobiotics are added to the media of both induced and uninduced cells. Initially, procarcinogens are applied in concentrations ranging from 1 to 50 .mu.M (19,20,67). In addition, the time course of exposure determined by using pulses of selected duration, starting with 1 hour. The time course data provide insight as to how quickly cytotoxic and genotoxic effects occur. After the pulse with procarcinogen, the cells are washed, harvested with trypsin and replated at lower density to allow for the formation of colonies. One set is incubated in the presence of 6-thioguanine to assay for mutation of the HPRT gene and one set is incubated in normal media to determine the cytotoxicity of the treatment.

e. Demonstration that CYP1A1 expressing cells metabolize BPD to cytotoxic agents.

Benzo(a)pyrene-trans-7,8-diol (BPD), Catalog No. L0113 was obtained from NCI Chemical Carcinogen Repository and a stock solution was prepared in tetrahydrofuran (THF) and stored at -20.degree. C. BPD dose response experience were performed with a number of clones, both CYP1A1 expressing and non-expressing, and parental cell lines in survival assays 25. Assays were performed in 24-well multiwell plates. Cell survival was measured by staining those cells still attached to the dish 24 to 48 hours after treatment. Cells were plated at 10.sup.4 per well and allowed to acclimate for 18 hours and the media was changed with fresh media. Half the wells received media containing 3 .mu.M CdSO.sub.4 (+Cd.sub.). BPDH in THF was added from 100.times. working stock solutions to the concentrations indicated below the bottom sets show in FIG. 9. It should be noted that the DNA repair proficient cells (right side of tray) were treated with 10-fold higher doses than were the XPA derived cells (left side of the tray). Control wells with no BPD received equivalent dose of THF and the cells were incubated 24 hours. At this time, the media was drained and the cells refed with fresh media containing not additives. After a further 24 to 48 hour incubation, the cells were washed, fixed with methanol, and stained with 0.04% methylene blue.

f. Demonstration that BPD cytotoxicity is prevented by .alpha.-NF

Although cadmium sulphate treatment of parental cells and neo.sup.+ cells without CYP1A1 does not induce BPD cytotoxicity, it is still essential to demonstrate that the cytotoxicity in CYP1A1 expressing cells is due to the metabolism of the BPD by the induced CYP1A1. Inclusion of .alpha.-NF an inhibitor of CYP1A1 activity, prevents the cytotoxicity of PBD in cadmium sulphate induced cells, the cell trays being shown in FIG. 7. Specifically, cells (clone 10) were plated at 10.sup.4 per well and allowed to acclimate for 18 hours and the media was changed with fresh media containing 3 .mu.M CDSO.sup.4. Half of the wells received .alpha.-NF (10 .mu.M final concentration, +.alpha.-NF) and half received acetone (-.alpha.-NF). BPD in THF was added from 100.times. working stock solutions to the concentrations indicated in FIG. 10. Control wells with no BPD received an equivalent dose of THF and the cells were incubated for 24 hours. At this time, the media was drained and the cells refed with fresh media containing no additives. After a further 24 to 48 hour incubation, the cells were washed, fixed with methanol, and stained with 0.4% methylene blue.

g. Demonstration that BPDE is equally cytotoxic under all conditions in all cell lines tested.

In contrast to the effects observed with BPD, BDPE was tested as to whether it was cytotoxic in all cell lines, regardless of cadmium sulphate treatment. Specifically, cells were plated and treated as the cells shown in FIG. 9, but with BPDE instead of BPD. The cell trays are shown in FIG. 11.

RESULTS

a. Characterization of XPA cells:

In addition to using the XPA cells for transient expression of normal and mutant human adenosine deaminase cDNAs (27), the XPA cells have been used as a tool for the isolation of a normal human cDNA for the DNA repair gene defective in XPA cells. The approach used consists of complementing XPA cells by co-transformation with a dominant selectable marker gene and a normal human cDNA expression library (28). Using PCR, stably integrated cDNAs from the genomic DNAs of complemented XPA cells have been isolated in an effort to identify the complementing sequences. Data obtained from these studies indicate that, like other human cells, the XPA cells to be used with the present invention integrate few copies of a transfected DNA and are unlikely to give high copy transformants. This is most useful for the production of a stable P450 chimeric gene copy number and a consistent expression level. Although the XPA cells integrate only a few copies of transfected DNAs (unpublished data), they do transform at high efficiency (19,28). As a result, transfection of these cells provides a large number of transformants which can be analyzed for copy number and expression.

b. Endogenous levels of cytochromes P450:

Most cultured cells do not express P450 after long term culture (3,29). However, it was vital to establish that the cells used indeed do not express P450, in particular CYP1A1. Therefore, the levels of CYP1A1 and cytochrome P450 IIE1 (CYP2E1) in GM4429 (XPA) and GM0637 (normal) fibroblasts were investigated by Western blot analyses using antibodies to the rat forms of these enzymes. No proteins corresponding to human CYP1A1 or CYP2E1 were detected by the antibodies. CYP1A1 is not normally present without induction. To ensure that the antibody to rat CYP1A1 would detect human CYP1A1, both XPA fibroblasts (GM4429) and HEPG2 cells (a human hepatoma cell line which expresses CYP1A1 when induced) were examined for the inducibility of CYP1A1 gene expression by a potent inducing agent, 2,3,7,8-tetrachlorodibenzofuran (TCDF) (FIG. 5).

FIG. 5 shows induction of CYP1A1 detected by western blot in HEPG2 cells but not in XPA fibroblasts. Cells were grown until approximately 75% confluent, then treated with 0.1 .mu.M 2,3,7,8-tetrachlorodibenzofuran (TCDF) or vehicle alone (ethanol). After 16 hours, the cells were harvested and microsomes were prepared. Microsomal proteins (10 .mu.g) from each cell line were fractionated by SDS-PAGE, transferred to nitrocellulose membrane and probed with antibody to rat CYP1A1. The CYP1A1 induced by TCDF in HEPG2 cells is visible as the darkest band in lane 2. No band having identical mobility is present in the other three lanes. The blot was deliberately overdeveloped to make certain there is no trace of CYP1A1 in the XPA fibroblasts. Microsomal proteins from: vehicle treated HEPG2 cells, lane 1: TCDF treated HEPG2 cells, lane 2; vehicle treated XPA fibroblasts, lane 3; TCDF treated XPA fibroblasts, cells, lane 4. The XPA fibroblasts did not show CYP1A1 expression induced by TCDF whereas the HEPG2 cells showed considerable induction of CYP1A1 expression as expected. Thus, the antibody for rat CYP1A1 cross reacts with similar determinants on human CYP1A1 as expected because the two proteins are highly homologous.

In addition, mRNAs from the induced and uninduced cells (both HEPG2 and XPA fibroblast) were analyzed by northern blot for CYP1A1 mRNA expression. FIG. 6 shows induction of CYP1A1 mRNA detected by northern blot in HEPG2 cells but not in XPA fibroblasts. Total RNA was prepared from cells treated concurrently with those of FIG. 1. The RNAs were fractionated by denaturing agarose gel electrophoresis, transferred to a modified nylon membrane and hybridized to a radiolabelled probe specific for human CYP1A1. CYP1A1 hybridizing mRNAs are visible as an intense band in the RNA from TCDF-treated HEPG2 cells (lane 2) but in none of the others. The autoradiograph was deliberately overexposed to make certain there is no trace of CYP1A1 mRNA in the XPA fibroblasts. RNAs from: vehicle-treated HEPG2 cells, lane 1; TCDF-treated HEPG2 cells, lane 2; vehicle-treated XPA fibroblasts, lane 3; TCDF-treated XPA fibroblasts, lane 4.

Thus, only the induced HEPG2 cells showed any detectable CYP1A1 mRNA. It was concluded that the human fibroblast cells do not express CYP1A1 or CYP2E1. Thus, they are suitable as recipients of P450 cDNA expression constructs.

In order to determine if the model system for assessing cytotoxicity and mutagenicity of metabolically activated xenobiotics will be useful, applicants chose to begin their studies with a paradigm of a xenobiotic that requires activation, benzo[a]pyrene, and its known bioactivator, CYP1A1.

c. Cloning and expression of human CYP1A1 coding sequences:

The human CYP1A1 gens was isolated from a Charon 4A genomic library and characterized by restriction endonuclease mapping and DNA sequence analysis (30). Several overlapping sub-clones have been constructed spanning from approximately 6000 bp 5' of the transcription initiation site to just 3' of the polyadenylation site (FIG. 4). FIG. 4 is a partial map of the 5830 bp human CYP1A1 gens. The open boxes represent non-coding exon information and closed boxes represent coding exon information. The position of restriction endonuclease recognition sites are shown for BglII (B), BclI (C), EcoRI (E), HinDIII (H), TaqI (T) and XbaI (X). The information contained in various subclones is shown below the map. The BclI/EcoRI fragment containing CYP1A1 coding information cloned into expression vectors to be used is denoted by the broken line. A shuttle vector, pRNH63, was prepared for the expression of the CYP1A1 structural gene under the direction of the mouse mammary tumor virus (MMTV) LTR and harvey murine sarcoma virus (HaMuSV) enhancer. Cloning of information into the unique BamH1 site of pRNH63 results in expression from the MTV/HaMuSV promoter/enhancer with 3' processing being contributed by the bovine growth hormone polyadenylation signal. In addition, kanamycin/Geneticin resistance, expressed under the prokaryotic P1 and eukaryotic tk promoters, provides a means for selection.

For the expression of CYP1A1 information, a 2.0 kb BclI/BamHI fragment containing the entire coding information for the CYP1A1 protein was cloned into the BamHI site of pRNH63 and used to transfect the human breast tumor cell line, MCF-7. Several stable transfectants were cloned. However, each exhibited an extremely slow growing phenotype suggesting that expression of the normally silent CYP1A1 gene under this relatively powerful promoter system was most likely toxic. Thus a less potent inducible promoter system is preferred for use with the present invention.

d. Choice of inducible promoter:

An inducible promoter allows direct comparison of expression to chemicals in the same cell line under conditions which either do or do not induce expression of the P450 cDNA. Using the same cells ensures that there will be no variation in expression due to rearrangements in multi-copy transgenes or position effects which can be encountered when using several different transformants. The mouse metallothionein-I (MMT-I) promoter is an inducible promoter with very low basal activity. As detailed below, its expression can be modulated by the concentration of metal ions in the media to give proportionately varied P450 levels. This is precisely what is needed to avoid the apparent toxicity of constitutive overproduction of CYP1A1 in human cells and will provide a means of correlating the level of P450 expression with mutagenic effect of a given dose of xenobiotic.

e. Studies using the MMT-I promoter:

The MMT-I promoter was tested in a transient assay system in the XPA fibroblasts (GM4429) using chloramphenicol acetyl transferase (CAT) as a reporter. MMT-I promoter potency was compared to that of several other promoters. The basal and induced activities of the MMT-I were compared with the activity of the human adenosine deaminase (ADA) promoter, the herpes simplex virus-1 thymidine kinase (HSV-tk) promoter and two strong, viral promoters. SV40 early promoter and Rous Sarcoma Virus LTR. The basal activity of the MMT-I promoter was found to be very similar to the activities of both the ADA and the HSV-tk promoters. These are very weak constitutive promoters. The MMT-I promoter showed proportional inducibility with increasing CdSO.sub.4 over the range of 1 to 5 .mu.M. Maximum induction of 32-fold was achieved with an overnight incubation with 5 .mu.M CdSO.sub.4 which was removed by washing the cells after the induction period. The cells seemed not to be adversely affected by this brief treatment and recovered well. Thus, the MMT-I promoter has very low activity in these cells but can be induced greatly by incubation with appropriate metals. If the low basal levels of the intact MMT-I promoter prove to be too high for use, a deletion mutant of the MMT-I promoter which has a 100-fold lower basal activity but is still inducible 20-fold by heavy metal can be used. This deletion mutant will surely give a basal activity which is low enough to avoid any problems associated with constitutive, high level expression of CYP1A1.

f. Synthesis and testing of CYP1A1 substrate:

Transformants were screened for inducible CYP1A1 activity by incubation of cells with ethoxy fluorescein ethyl ester (EFEE, see Experimental Methods). EFEE is not available commercially. Therefore, it is necessary to synthesize EFEE. The procedure of Miller (31) as modified by Salata (21) was used to prepare sufficient EFEE to screen several thousand cultures for CYP1A1 activity. The EFEE synthesized has been tested with HepG2 cells induced for CYP1A1 expression by TCDF exposure (FIG. 7).

HepG2 cells were plated in a 24-well multiwell plate and incubated until approximately 75% confluent. CYP1A1 was induced in the cells in half the wells (lanes 2 & 4) by addition of TCDF to 10 .mu.M and the cells were incubated for 24 hours. The cells were washed three times with PBS, overlaid with 100 .mu.l PBS containing EFEE at 500 nM (lanes 1 & 2) or 5 .mu.M (lanes 3 & 4) and incubated for one hour. The PBS-EFEE from each well was withdrawn and acidified by addition of 10 .mu.l 1N HCl. Thin layer chromatography was performed with 50 .mu.l aliquots of the acidified samples. The chromatograph was developed with ethylacetate:benzene (3:1) and visualized by illumination with long wave UV. At lower doses of EFEE (500 nM), nearly all is converted to FEE by the HepG2 cells. At a higher dose (5 .mu.M), only a portion of the EFEE is converted to FEE. Thus, applicants have successfully synthesized this critical reagent and demonstrated its usefulness in screening small scale cultures for inducible CYP1A1 activity both rapidly and easily.

g. Characterization of CYP1A1 Transformed Fibroblasts:

DNA repair normal (GM0637) and XPA (GM4429 & GM4312) fibroblasts were co-transfected with the dominant selectable marker pRSV-NEO and the CYP1A1 chimeric gene construct at three different ratios of DNAs as described in Experimental Methods. Several hundred colonies of NEO+ transformants were obtained from one XPA (GM4429) and the normal (GM0673) transfected fibroblasts by plating at low density and incubating in media containing G-418 (300 .mu.g/ml). Twenty-three colonies were isolated and grown individually in mass culture. Analysis for CYP1A1 activity inducible by CdSO.sub.4 was performed utilizing the EFEE assay. One transformant (of 23 tested) was identified as having CYP1A1 activity inducible by 5 82 M CdSO.sub.4 (FIG. 8).

Individual NEO+ transformant colonies were isolated and grown in mass culture. Cells were trypsinized and plated in duplicate in 24-well multi-well plates. When the cells were -75% confluent, the media was changed. One well of each pair received 0.5 ml normal media and the other received 0.5 ml media containing 5 .mu.M CdSO.sub.4. After incubation overnight, the media was removed, the cells were washed with PBS and then overlaid with 100 .mu.l PBS containing 2.5 .mu.M EFEE and incubated 1.5 hours at 37.degree. C. The PBS-EFEE from each well was withdrawn and acidified by addition of 10 .mu.l 1 N HCl. Thin layer chromatography was performed with 50 .mu.l aliquots of the acidified samples. The chromatograph was developed with ethylacetate:benzene (3:1) and visualized by illumination with long wave UV. Lanes 1 & 2 are clone 6C cells; lanes 3 & 4 are clone 4T cells. Lanes 1 & 3 are cells induced with 5 .mu.M CdSO.sub.4. The transformant expressing CYP1A1 (clone 6C) was XPA (GM4429) cells transfected with a 10:1 ratio of CYP1A1:pRSV-NEO DNAs and is clone 6C. All other transformants analyzed had no CYP1A1 activity (like clone 4T). The CYP1A1 activity inducible in clone 6C cells is considerably lower than that inducible by TCDF in HepG2 cells (FIG. 7). The CYP1A1 activity in uninduced clone 6C cells is undetectable. These data indicate that XPA fibroblasts transformed with a CYP1A1 chimeric gene construct that uses the mouse metallothionein-I promoter to drive transcription of the CYP1A1 sequences and that is inducible by incubation with 5 .mu.M CdSO.sub.4 can be obtained.

h. Demonstration of utilizing CYP1A1 expressing cells for cytotoxicity tests.

Typical results of benzo(a)pyrene-trans-7,8-diol (BPD) dose response assays are shown in FIG. 9. In the experiment, the +Cd cells were exposed to 3 .mu.M CdSO.sub.4 for 24 hours. Similar results were obtained with 7.mu. CdSO.sub.4 for six hours. The higher dose of CdSO.sub.4 induces high levels of CYP1A1 and allows for faster metaboloism of BPD such that an equivalent amount of the end product is produced, as shown in the Figure. This brief exposure to a slightly higher CDSO.sub.4 concentration also does not induce toxic effects.

As shown in FIG. 9, BPD up to 10 .mu.M was not cytotoxic parental cells (GM4429 of GM0637).+-.cadmium sulphate. BPD up to 10 .mu.M was also not cytotoxic to transformants not expressing CYP1A1 (6D or 120.+-.CdSO.sub.4) or to uninduced CYP1A1 transformants (10 or 8D, both without CdSO.sub.4). In contrast, CYP1A1 expressing cells (10 and 8D both plus CdSO.sub.4) showed increasing cytotoxicity with increasing concentration of BPD indicated by the loss of staining cells with increasing BPD dose. Furthermore, a >10-fold differential in the cytotoxicity between 10 cells and 8D cells is observed. XPA derived 10 cells display 100% cytotoxicity with 0.1 .mu.M BPD whereas the DNA repair normal 8D cells can withstand more than 1 .mu.M BPD before complete cytotoxicity is observed. These data indicate that CYP1A1 expressing cells metabolize BPD into DNA damaging agent that causes bulky adduct damage to the DNA which the XPA cells cannot repair. These results are exactly what is predicted based on the presumed role of CYP1A1 in the bioactivation of BPD into BPDE. These results suggest that the Cd induced CYP1A1 expressing DNA repair deficient and proficient cells can be used to detect proximate DNA damaging agents activated by CYP1A to ultimate DNA damaging agents.

i. Demonstration that cytotoxicity in CYP1A1 expressing cells is due to metabolism of BPD by the induced CYP1A1.

As shown in FIG. 10, inclusion of .alpha.-napthoflauore (.alpha.-NF), an inhibitor of CYP1A1 activity, prevented the cytotoxicity of BPD in CdSO.sub.4 induced 10 cells. This result suggests that the cytotoxicity caused by BPD in CYP1A1 expressing cells is due to metabolism of the BPD by the CYP1A1 and supports the conclusion that the CYP1A1 transformants metabolize BPD in situ to a potent DNA damaging agent.

j. Demonstration that BPDE is equally cytotoxic under all conditions in all cell lines.

In contrast to the effects observed with BPD, BPDE (benzo(a)pyrene-trans-7,8-diol-9,10-expoxide) is equally cytotoxic in all the cell lines regardless of CdSO.sub.4 treatment (FIG. 11). The lack of differential cytotoxicity between XPA derived (GM4429, 10 & 6D) and DNA repair normal (GM0637, 8D, 120) cells indicates that the major mechanism(s) of cytotoxicity due to externally added BPDE does not include DNA damage events. It is likely that the relatively reactive diol epoxide is reacting with cell membrane components and other essential macromolecules such as RNA and proteins prior to arriving in the nucleus. BPDE metabolically produced in situ is synthesized in the endoplasmic reticulum. The interior of the endoplasmic reticulum is topologically contiguous with the internal milieu of the nucleus affording relatively direct access to the genomic material. Therefore, BPDE produced in the endoplasmic reticulum is less likely to encounter RNAs and cell membrane components than is BPDE applied externally to the cells in the media. Thus, metabolism of BPD in situ is a more specific means of inducing DNA damage by BPDE than is exposure of cells to BPDE in the media. Additionally, the potential effects of adduction of cellular macromolecules on DNA repair capability in studies of DNA repair and mutagenesis must be considered. Cells treated with BPDE in the media have suffered a wide range of insult and many components of the cellular machinery have been damaged.

In conclusion, the experiments performed to date demonstrate that human fibroblasts, both DNA repair deficient and proficient, have been stably transformed with a controllable CYP1A1 expression construct. Treatment of these cells with BAP metabolite BPD, while expression of CYP1A1 is induced, results in the conversion of BPD into a potent DNA damaging agent(s) that elicits a differential cytotoxicity with DNA repair deficient (XPA derived) and DNA repair normal cells. These CYP1A1 expressing cells distinguish subtle, but vitally important, differences in the cytotoxic and genotoxic effects between BPDE produced metabolically in situ and BPDE applied externally to the cells.

The invention has been described in an illustrative manner, and it is to be understood that the terminology which has been used is intended to be in the nature of words of description rather than of limitation.

Obviously many modifications and variations of the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims the invention may be practiced otherwise than as specifically described.

    __________________________________________________________________________
    SEQUENCE LISTING
    (1) GENERAL INFORMATION:
    (iii) NUMBER OF SEQUENCES: 2
    (2) INFORMATION FOR SEQ ID NO:1:
    (i) SEQUENCE CHARACTERISTICS:
    (A) LENGTH: 6387 base pairs
    (B) TYPE: nucleic acid
    (C) STRANDEDNESS: double
    (D) TOPOLOGY: circular
    (ii) MOLECULE TYPE: DNA (genomic)
    (vi) ORIGINAL SOURCE:
     (A) ORGANISM: Homo sapiens
    (ix) FEATURE:
    (A) NAME/KEY: exon
    (B) LOCATION: complement (2807..3631)
    (ix) FEATURE:
    (A) NAME/KEY: exon
    (B) LOCATION: complement (2125..2251)
    (ix) FEATURE:
    (A) NAME/KEY: exon
    (B) LOCATION: complement (1948..2037)
    (ix) FEATURE:
    (A) NAME/KEY: exon
     (B) LOCATION: complement (1733..1856)
    (ix) FEATURE:
    (A) NAME/KEY: exon
    (B) LOCATION: complement (1501..1587)
    (ix) FEATURE:
    (A) NAME/KEY: exon
    (B) LOCATION: complement (237..1308)
    (ix) FEATURE:
    (A) NAME/KEY: promoter
    (B) LOCATION: complement (3638..3967)
    (ix) FEATURE:
    (A) NAME/KEY: CDS
     (B) LOCATION: 4586..5446
    (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:
    GCGCCCAATACGCAAACCGCCTCTCCCCGCGCGTTGGCCGATTCATTAATGCAGCTGGCA60
    CGACAGGTTTCCCGACTGGAAAGCGGGCAGTGAGCGCAACGCAATTAATGTGAGTTAGCT120
    CACTCATTAGG CACCCCAGGCTTTACACTTTATGCTTCCGGCTCGTATGTTGTGTGGAAT180
    TGTGAGCGGATAACAATTTCACACAGGAAACAGCTATGACCATGATTACGAATTCGACAA240
    ACATTAAAAATAGACATTTTATTACAAGAGTGTAGAGAAGGGAGACCAATAGAAG GTAAT300
    TGAAATACCCCCCCCTCACTCCAGCCCTAGTCCTGGTGCCTGGATATGTGCACTCCCTGT360
    GCGCTCTGATCCCCGCAGACACAAGTCCCCAGCCCCTCCAGGACAGCAATAAGGGTCTTA420
    CAAGGCCAGAAGGCAGCCCTGTTTGTTCCT GCCTGCAGGAAGGGCAGAGGAATGTGATGT480
    TCCCAGGAACTGTGTCCTAGACCCATAGGGTCAGATTGCTCAGCCTAGTTCAAGCAGTGA540
    GACTACCTCTGTGCCAGTATCCTGGGCTGTCTCTTCCCTTCACTCTTGGCAGCTCCCAAT600
    TGTC AAAGATTGGACAGGGTCCTGGTTTGGCTAGTTCTAACTTGCTGAAGCCAGTCAGCA660
    CCCTCACAGAGCCAGCTAGGTACTGGGCCCAGGGGCTTCCAGAGAGTTCTTCAGAGCTTC720
    TCAGAGGCCTAAGGACCTCCTAACCCTAGCAGGCCTCCTGGCTCAAGC ACAACTTGGGAA780
    GGCTCCATCAGCATCTATGTGGCCCTGTTTTACCTGTTGTCTCTGGAGGGTGTGCAGAGG840
    CAAGTCCAGGGTAGGGGCAGGCAGGATCCCTTAGGCTTGCCCACAGCCCAGATAGCAAAA900
    CTGCAGCCAGATCAGTGTCTAT GAGTTTCAGGCTGAATCTTAGACCACATAGGCCAGCCT960
    GCTGGTCTGGCTGCCCAACCAGACCAGGTAGACAGAGTCTAGGCCTCAGGGCTCTCAAGC1020
    ACCTAAGAGCGCAGCTGCATTTGGAAGTGCTCACAGCAGGCATGCTTCATGGTTAGCCCA1080
    TAGATGGGGGTCATGTCCACCTTCACGCCCAGTGGCACGCTGAATTCCACCCGTTGCAGC1140
    AGGATAGCCAGGAAGAGAAAGACCTCCCAGCGGGCAATGGTCTCACCGATACACTTCCGC1200
    TTGCCCATGCCAAAGATAATCACCTTCTCACTTAACACCT TGTCGATAGCACCATCAGGG1260
    GTGAGAAACCGTTCAGGTAGGAACTCAGATGGGTTGACCCATAGCTTCCTGTAACCAGAG1320
    GGAGACAGCTGAAGTGGCAGTTCAGGGCTCAGAAGTGTCAAGTGAGTGGAGCTCCAGCCC1380
    CAAAGGATAGAGGAC AGGCAAGCAGCCCATGGACAGGAGGATCAATGCAATGATTGTATT1440
    AATCATATATAAGAGCTTAAGAGGGTGGACCCAGCCTTTCCTCTGCATCTCTGAACTTAC1500
    TGGTCATGGTTGATCTGCCACTGGTTTACAAAGACACAACGCCCCTTGGGGATGTAAAA G1560
    CCTTTCAAACTTGTGTCTCTTGTTGTGCTAGGGAGAAAGGAAGCTCAGTCAGGCTCAGGG1620
    CAACAGGCAAATCTCCCTGTCTCCCATGCCGTGTCCCTCCCACTAACCCTAATCAGGTAT1680
    GTGGTCCGGAGTAAGATCAGTAACAGACAGCAG TGGCTCCATGGGGCCTTACCTGTGGGG1740
    GATGGTGAAGGGGACGAAGGAAGAGTGTCGGAAGGTCTCCAGGATGAAGGCCTCCATATA1800
    GGGCAGATGGGATCTGTCAGAGAGCCGGGGCCGCCGTGACCTGCCAATCACTGTGTCTGC1860
    AGAACACA GGGACAAGATGGATGCAGGGGCTGCCTAGACCTGGCCAGACCCCTGGCACTG1920
    ACCCCTTTGAAGGGAGCCACTACCTACCTAGCTCCTCTTGGATCTTTCTCTGTACCCTGG1980
    GGTTCATCACCAAATACATGAGGCTCCAGGAGATAGCAGTTGTGACTGTGT CAAACCCTG2040
    GACAGGGTAGAACAGAAGAAGTTAGGCAGGCAGCAGCAGGTCAGGGCACTTGAGCACAGG2100
    AAGGACACAATGGGGTAACCATACCAGCTCCAAAGAGGTCCAAGACGATGTTAATGATCT2160
    TCTCATCTGACAGCTGGACATTGGCG TTCTCATCCAGCTGCTTCTCCTGACAGTGCTCAA2220
    TCAGGCTGTCTGTGATGTCCCGGATGTGGCCCTTAGGTAGGGAAAGTCCACAGGTGAGCA2280
    AGATCTCAAACCCAGAGCTACCTCTCCATCCAGGTCTGGTCCTTCACTATTCCTAGCACA2340
     TTTGTTCTGGAGGTGATGCCCCCTGAGGCTGTTGTCCCAGCTTCTCTCTGCCTCTGCAAG2400
    GCTCTCTCCTACTACCTTAGAATGCTGCTAGCCCCAACTCATGGGACATTTGACACACAG2460
    CGTCTTGTACTGTTATCATCTGGATGTGCTCTTTATCTTCTTTC TATTATGAGGCCAGGA2520
    GATGAACTCTTTAACTCTTCCTGGCTCCCTCAGCAACTGCCCCAGGGTCCTGCATGTAAT2580
    GACTCTTCAGTGGCTATTGCTGTCTGTGGAAGCATGGAAGGGTTAGTCAAGATAAAGTTC2640
    TATTTCCCTGCCAAGGAAG AAGACTATTCCACAACTGGCTTCAAGATCCCAGGTTGAAGC2700
    CTTCCTGAGAACTTGCCAAGCCCCATGCAGTTCCTCTTACCTTTGACCTCCCAGGCCCTG2760
    ATGCCATCTGCTTCCCACCACCCACCTGCCTTTCCCCAGACTGTACCTTCTCAAAGGTTT 2820
    TGTAGTGCTCCTTGACCATCTTCTGCATGAAGCTGTAGAACTTCTCATTCAGGTCCTTGA2880
    AGGCATTCAGGGAAGGGTTGGGTAGGTAGCGAAGAATAGGGATGAAGTCAGCTGGGTTTC2940
    CAGAGCCAACCACCTCCCCGAAATTATTATTCAGGTT GACTAGGCTAAGCAGTTCTTGGT3000
    GGTTGTGGTCATAGCGCCGGCCAAAGCAAATGGCACAGATGACATTGGTCACTGATACCA3060
    CCACATACCTGTAGGGGTTAAAGTGCCCAGGCCCTGCCATCAGCTCCTGCAACGTGCTTA3120
    TCAGGACCTCA GCCTCCTTGCTCACATGCTCTTCCAGGTAGCAGGAGGTTGAGGAGGCTG3180
    GGTCAGAGGCAATGGAGAAACTTTTCAGGCCATTCTGGGCCAGGCGCCGGCGGGCAGCCC3240
    ACACTGGTCCAGAGTCTGGGCTGAAGGACATGCTCTGACCATTACTGATGAGGGT GAAGG3300
    TGTAGAGGTCGGGCCGGCCCTTGAAATCATCGCCCTGCCGCACCAGGGCCTGCCGGATGG3360
    TGTCCAGGCCGCTCAGCACCACCACGGGTGTGGAGCCAATTCGGATCTGCAGCACGTCCC3420
    CATACTGCTGGCTCATCCTTGACAGTGCCA GGTGCGGGTTCTTTCCCAGGGTCAGCATGT3480
    GCCCAATCAGAGGCCAGCCCCATGGCCCTGGTGGATTCTTCAGGCCTTTGGGGACCTGAG3540
    GTCTTGAGGCCCTGATTACCCAGAATACCAGACAGAAGATGACAGAGGCCAGAAGAAACT3600
    CCGT GGCCGACATGGAGATTGGGAAAAGCATGATCTGGTGAAGCTGGAGCTACGGAGTAA3660
    GTGAGGAGAAGGTACTCAGGACGTTGAAGTCGTGGTGACGCTTAGAGGACAGCCTGCCCT3720
    CTTTATAGTCGTTGGACGAGTCCGGGCGCAAAGGGTTTGCACCCAGCA GGCGGTTGCTCC3780
    AGCCCACGCATAGTCACGCGCCCCGCGTCCTGGCAGAGCCGAGCGCACTTTTCGGGCGGA3840
    GTGCAGAGCTCCCTGGAGCGCCAGTGTGCACAGCGGGCACGCCCCTTTGGCACGACTGGC3900
    TCGGATAGGACGACCCATGTGA CGTGTGGAACATCATGTCCCTATAGTGCTTTCCCCGAT3960
    ATTACGGGGGCGAGCTCGAATTAGCTTGGCACTGGCCGTCGTTTTACAACGTCGTGACTG4020
    GGAAAACCCTGGCGTTACCCAACTTAATCGCCTTGCAGCACATCCCCCTTTCGCCAGCTG4080
    GCGTAATAGCGAAGAGGCCCGCACCGATCGCCCTTCCCAACAGTTGCGCAGCCTGAATGG4140
    CGAATGGCGCCTGATGCGGTATTTTCTCCTTACGCATCTGTGCGGTATTTCACACCGCAT4200
    ATGGTGCACTCTCAGTACAATCTGCTCTGATGCCGCATAG TTAAGCCAGCCCCGACACCC4260
    GCCAACACCCGCTGACGCGCCCTGACGGGCTTGTCTGCTCCCGGCATCCGCTTACAGACA4320
    AGCTGTGACCGTCTCCGGGAGCTGCATGTGTCAGAGGTTTTCACCGTCATCACCGAAACG4380
    CGCGAGACGAAAGGG CCTCGTGATACGCCTATTTTTATAGGTTAATGTCATGATAATAAT4440
    GGTTTCTTAGACGTCAGGTGGCACTTTTCGGGGAAATGTGCGCGGAACCCCTATTTGTTT4500
    ATTTTTCTAAATACATTCAAATATGTATCCGCTCATGAGACAATAACCCTGATAAATGC T4560
    TCAATAATATTGAAAAAGGAAGAGTATGAGTATTCAACATTTCCGTGTCGCC4612
    MetSerIleGlnHisPheArgValAla
    15
    CTTATTCCCTTTTTTGCGGCATTTTGCCTTCCTGTTTTTGCTCACCCA4660
    LeuIleProPhePheAlaAlaPheCysLeuProValPheAlaHisPro
    101520 25
    GAAACGCTGGTGAAAGTAAAAGATGCTGAAGATCAGTTGGGTGCACGA4708
    GluThrLeuValLysValLysAspAlaGluAspGlnLeuGlyAlaArg
    3035 40
    GTGGGTTACATCGAACTGGATCTCAACAGCGGTAAGATCCTTGAGAGT4756
    ValGlyTyrIleGluLeuAspLeuAsnSerGlyLysIleLeuGluSer
    4550 55
    TTTCGCCCCGAAGAACGTTTTCCAATGATGAGCACTTTTAAAGTTCTG4804
    PheArgProGluGluArgPheProMetMetSerThrPheLysValLeu
    6065 70
    CTATGTGGCGCGGTATTATCCCGTATTGACGCCGGGCAAGAGCAACTC4852
    LeuCysGlyAlaValLeuSerArgIleAspAlaGlyGlnGluGlnLeu
    758085
    GGTCGCCGCATACACTATTCTCAGAATGACTTGGTTGAGTACTCACCA4900
    GlyArgArgIleHisTyrSerGlnAsnAspLeuValGluTyrSerPro
    9095100 105
    GTCACAGAAAAGCATCTTACGGATGGCATGACAGTAAGAGAATTATGC4948
    ValThrGluLysHisLeuThrAspGlyMetThrValArgGluLeuCys
    110115 120
    AGTGCTGCCATAACCATGAGTGATAACACTGCGGCCAACTTACTTCTG4996
    SerAlaAlaIleThrMetSerAspAsnThrAlaAlaAsnLeuLeuLeu
    125130 135
    ACAACGATCGGAGGACCGAAGGAGCTAACCGCTTTTTTGCACAACATG5044
    ThrThrIleGlyGlyProLysGluLeuThrAlaPheLeuHisAsnMet
    140145 150
    GGGGATCATGTAACTCGCCTTGATCGTTGGGAACCGGAGCTGAATGAA5092
    GlyAspHisValThrArgLeuAspArgTrpGluProGluLeuAsnGlu
    155160165
    GCCATACCAAACGACGAGCGTGACACCACGATGCCTGTAGCAATGGCA5140
    AlaIleProAsnAspGluArgAspThrThrMetProValAlaMetAla
    170175180 185
    ACAACGTTGCGCAAACTATTAACTGGCGAACTACTTACTCTAGCTTCC5188
    ThrThrLeuArgLysLeuLeuThrGlyGluLeuLeuThrLeuAlaSer
    190195 200
    CGGCAACAATTAATAGACTGGATGGAGGCGGATAAAGTTGCAGGACCA5236
    ArgGlnGlnLeuIleAspTrpMetGluAlaAspLysValAlaGlyPro
    205210 215
    CTTCTGCGCTCGGCCCTTCCGGCTGGCTGGTTTATTGCTGATAAATCT5284
    LeuLeuArgSerAlaLeuProAlaGlyTrpPheIleAlaAspLysSer
    220225 230
    GGAGCCGGTGAGCGTGGGTCTCGCGGTATCATTGCAGCACTGGGGCCA5332
    GlyAlaGlyGluArgGlySerArgGlyIleIleAlaAlaLeuGlyPro
    235240245
    GATGGTAAGCCCTCCCGTATCGTAGTTATCTACACGACGGGGAGTCAG5380
    AspGlyLysProSerArgIleValValIleTyrThrThrGlySerGln
    250255260 265
    GCAACTATGGATGAACGAAATAGACAGATCGCTGAGATAGGTGCCTCA5428
    AlaThrMetAspGluArgAsnArgGlnIleAlaGluIleGlyAlaSer
    270275 280
    CTGATTAAGCATTGGTAACTGTCAGACCAAGTTTACTCATATATACTTTAGATTG5483
    LeuIleLysHisTrp
    285
    ATTTAAAACTTCATTTTTAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGATAATCTCA 5543
    TGACCAAAATCCCTTAACGTGAGTTTTCGTTCCACTGAGCGTCAGACCCCGTAGAAAAGA5603
    TCAAAGGATCTTCTTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGCTTGCAAACAAAAA5663
    AACCACCGCTACCAGCGGTGGTTTGTTTGCCGGATCA AGAGCTACCAACTCTTTTTCCGA5723
    AGGTAACTGGCTTCAGCAGAGCGCAGATACCAAATACTGTCCTTCTAGTGTAGCCGTAGT5783
    TAGGCCACCACTTCAAGAACTCTGTAGCACCGCCTACATACCTCGCTCTGCTAATCCTGT5843
    TACCAGTGGCT GCTGCCAGTGGCGATAAGTCGTGTCTTACCGGGTTGGACTCAAGACGAT5903
    AGTTACCGGATAAGGCGCAGCGGTCGGGCTGAACGGGGGGTTCGTGCACACAGCCCAGCT5963
    TGGAGCGAACGACCTACACCGAACTGAGATACCTACAGCGTGAGCTATGAGAAAG CGCCA6023
    CGCTTCCCGAAGGGAGAAAGGCGGACAGGTATCCGGTAAGCGGCAGGGTCGGAACAGGAG6083
    AGCGCACGAGGGAGCTTCCAGGGGGAAACGCCTGGTATCTTTATAGTCCTGTCGGGTTTC6143
    GCCACCTCTGACTTGAGCGTCGATTTTTGT GATGCTCGTCAGGGGGGCGGAGCCTATGGA6203
    AAAACGCCAGCAACGCGGCCTTTTTACGGTTCCTGGCCTTTTGCTGGCCTTTTGCTCACA6263
    TGTTCTTTCCTGCGTTATCCCCTGATTCTGTGGATAACCGTATTACCGCCTTTGAGTGAG6323
    CTGA TACCGCTCGCCGCAGCCGAACGACCGAGCGCAGCGAGTCAGTGAGCGAGGAAGCGG6383
    AAGA6387
    (2) INFORMATION FOR SEQ ID NO:2:
    (i) SEQUENCE CHARACTERISTICS:
    (A) LENGTH: 286 amino acids
     (B) TYPE: amino acid
    (D) TOPOLOGY: linear
    (ii) MOLECULE TYPE: protein
    (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:
    MetSerIleGlnHisPheArgValAlaLeuIleProPhePheAlaAla
    151015
    PheCys LeuProValPheAlaHisProGluThrLeuValLysValLys
    202530
    AspAlaGluAspGlnLeuGlyAlaArgValGlyTyrIleGluLeuAsp
    35 4045
    LeuAsnSerGlyLysIleLeuGluSerPheArgProGluGluArgPhe
    505560
    ProMetMetSerThrPheLysValLeuLeuCys GlyAlaValLeuSer
    65707580
    ArgIleAspAlaGlyGlnGluGlnLeuGlyArgArgIleHisTyrSer
    8590 95
    GlnAsnAspLeuValGluTyrSerProValThrGluLysHisLeuThr
    100105110
    AspGlyMetThrValArgGluLeuCysSerAlaAlaIleThrM etSer
    115120125
    AspAsnThrAlaAlaAsnLeuLeuLeuThrThrIleGlyGlyProLys
    130135140
    GluLeuThrAlaPhe LeuHisAsnMetGlyAspHisValThrArgLeu
    145150155160
    AspArgTrpGluProGluLeuAsnGluAlaIleProAsnAspGluArg
    165 170175
    AspThrThrMetProValAlaMetAlaThrThrLeuArgLysLeuLeu
    180185190
    ThrGlyGluLeuLeuThrLeuAla SerArgGlnGlnLeuIleAspTrp
    195200205
    MetGluAlaAspLysValAlaGlyProLeuLeuArgSerAlaLeuPro
    2102152 20
    AlaGlyTrpPheIleAlaAspLysSerGlyAlaGlyGluArgGlySer
    225230235240
    ArgGlyIleIleAlaAlaLeuGlyProAspGlyLysProSerArgI le
    245250255
    ValValIleTyrThrThrGlySerGlnAlaThrMetAspGluArgAsn
    260265270
    ArgGln IleAlaGluIleGlyAlaSerLeuIleLysHisTrp
    275280285
    __________________________________________________________________________



Top